Changes in Microbiota in Type II Diabetes Patients

Anita Skovbjerg Hjort-Gregersen - s201190

Asiya Mohamad Yusuf Muses - 243930

Magnus Harthimmer - s233426

Marco Andreis - s243116

Rebecca Hjermind Millum - s215024

2024-12-03

Humans as a complex ecosystems

key point 1
key point 2
key point 3

Data set history

8 human associated shot-gun metagenomics studies
Consolidated for assessment of relative abundance
Converted to CSV and posted on Kaggle

Diagram of data processing

Data selection

Selecting the WT2D data set
Data processing

The WT2D dataset

Hallmarks of Diabetes

HDL (“Good Cholesterol”)

Lower in Type 2 Diabetes
Risk: Heart disease, stroke
Factors: Medications, overweight
Distribution: Shift to lower values

Hemoglobin A1C

Measures: Avg. glucose (10–12 weeks)
Threshold: >48 mmol/mol = Type 2 Diabetes
Findings: 34% of Type 2 Diabetes cases above threshold

BMI

Not Significant
Waist Circumference: Stronger correlation

HDL Notes

HDL, commonly referred to as “good” cholesterol, is known to be lower in individuals with type 2 diabetes. This reduction increases the risk of heart disease and stroke. Other documented factors that contribute to lower HDL levels include certain medications used to treat type 2 diabetes and being overweight. In our analysis, HDL was identified as a statistically significant parameter. The distribution plot shows a clear shift to the left, indicating lower HDL values in this group.

Relevant image: 07_ridges_hdl_vs_glucose_tolerance.jpg

Hemoglobin A1C Notes

Hemoglobin A1C is a key measure that reflects the average plasma glucose concentration over the previous 10–12 weeks. A value above 48 mmol/mol is a strong indicator of type 2 diabetes. In our analysis, Hemoglobin A1C was also statistically significant. The distribution plot highlights that 34% of individuals with type 2 diabetes had values exceeding 48 mmol/mol, whereas the other two groups had 0% above this threshold.

Relevant images: - 07_hba1c_vs_glucose_tolerance.jpg - 07_hba1c_48_threshold.jpg

BMI Notes

BMI is often considered a statistically significant predictor of diabetes. However, in our study, we did not observe the same correlation. Instead, waist circumference emerged as a noteworthy parameter. These two measures are somewhat correlated, as shown in the analysis.

Relevant image: 07_bmi_vs_glucose_tolerance.jpg

Hallmark of Type 2 Diabetes

GAD-Antibodies

Associated with Type 1 Diabetes
Pattern consistent across BMI classes

C-Peptide

Marker for insulin production
Higher in Type 2 Diabetes

Fasting Insulin

Higher in Type 2 Diabetes
Statistically significant parameter

GAD Antibodies Notes

GAD (Glutamic Acid Decarboxylase) antibodies are typically associated with type 1 diabetes. As expected, this parameter is not statistically significant in our analysis of type 2 diabetes. Additionally, this pattern holds consistently across different BMI classes, reinforcing its lack of relevance in type 2 diabetes contexts.

Relevant image: 07_gad_ab_vs_bmi_glucose_tolerance.jpg

C-Peptide Notes

C-peptide is a 31-amino-acid sequence that connects the A and B chains of proinsulin. It is cleaved during insulin production, making its levels a key marker to differentiate between insulin-producing (type 2 diabetes) and non-producing (type 1 diabetes) patients. Higher levels of C-peptide in type 2 diabetes patients indicate that the body is producing more insulin to compensate for elevated blood sugar levels.

Relevant image: 07_c_peptide_vs_bmi_glucose_tolerance.jpg

Fasting Insulin Notes

As expected, patients with type 2 diabetes tend to have higher levels of fasting insulin due to insulin resistance. This was confirmed in our analysis, where fasting insulin was identified as a statistically significant parameter.

Relevant image: 07_fasting_insuling_vs_glucose_tolerance_bar.jpg

Changes in Microbiota in Type II Diabetes Patients

Humans as a complex ecosystems

Data set history

Data selection

The WT2D dataset

Hallmarks of Diabetes

Hallmark of Type 2 Diabetes

Hallmarks of Type II Diabetes

Analysing the abundances

Statins - How do they interact?

Does diabetes change the microbiota?

Conclusions